Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918509983825920 |
|---|---|
| author | Tang, Zecheng Fu, Jiaye Gao, Qiankun Li, Haijie Wu, Yanmin Zhang, Jiaqi Ma, Siwei Zhang, Jian |
| author_facet | Tang, Zecheng Fu, Jiaye Gao, Qiankun Li, Haijie Wu, Yanmin Zhang, Jiaqi Ma, Siwei Zhang, Jian |
| contents | Feed-forward 3D reconstruction models based on Vision Transformers can directly estimate scene geometry and camera poses from a small set of input images, but scaling them to video inputs with hundreds or thousands of frames remains challenging due to the quadratic cost of global attention layers. Recent token-merging methods accelerate these models by compressing the token sequence within the global attention layers, but they apply a uniform reduction to query tokens and key-value tokens, ignoring their functionally distinct roles in 3D reconstruction. In this work, we identify a key property of feed-forward 3D reconstruction models: query tokens encode view-specific geometric requests and are sensitive to compression, while key-value tokens represent shared scene context and tolerate aggressive compression. Guided by this insight, we propose Spark3R, a training-free acceleration framework that decouples the compression of query tokens and key-value tokens by assigning distinct reduction factors, with intra-group token merging applied to query tokens and lightweight token pruning to key-value tokens. Additionally, Spark3R adaptively adjusts the key-value reduction factor across layers, further improving the quality-efficiency trade-off. As a plug-and-play framework requiring no retraining, Spark3R integrates directly into multiple pretrained feed-forward 3D reconstruction models, including VGGT, $π^3$, Depth-Anything-3, and VGGT-$Ω$, and achieves up to $28\times$ speedup on 1,000-frame inputs while maintaining competitive reconstruction quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_06270 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction Tang, Zecheng Fu, Jiaye Gao, Qiankun Li, Haijie Wu, Yanmin Zhang, Jiaqi Ma, Siwei Zhang, Jian Computer Vision and Pattern Recognition Feed-forward 3D reconstruction models based on Vision Transformers can directly estimate scene geometry and camera poses from a small set of input images, but scaling them to video inputs with hundreds or thousands of frames remains challenging due to the quadratic cost of global attention layers. Recent token-merging methods accelerate these models by compressing the token sequence within the global attention layers, but they apply a uniform reduction to query tokens and key-value tokens, ignoring their functionally distinct roles in 3D reconstruction. In this work, we identify a key property of feed-forward 3D reconstruction models: query tokens encode view-specific geometric requests and are sensitive to compression, while key-value tokens represent shared scene context and tolerate aggressive compression. Guided by this insight, we propose Spark3R, a training-free acceleration framework that decouples the compression of query tokens and key-value tokens by assigning distinct reduction factors, with intra-group token merging applied to query tokens and lightweight token pruning to key-value tokens. Additionally, Spark3R adaptively adjusts the key-value reduction factor across layers, further improving the quality-efficiency trade-off. As a plug-and-play framework requiring no retraining, Spark3R integrates directly into multiple pretrained feed-forward 3D reconstruction models, including VGGT, $π^3$, Depth-Anything-3, and VGGT-$Ω$, and achieves up to $28\times$ speedup on 1,000-frame inputs while maintaining competitive reconstruction quality. |
| title | Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.06270 |